INDEX

Explanations

mentions of the term "bot" in the text

oai_token-act-pair · gpt-3.5-turbo

mentions of robots or automated entities

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.3.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

journal

-0.76

mble

-0.75

uncture

-0.74

egal

-0.68

eway

-0.62

ournal

-0.59

VEN

-0.58

 ACTION

-0.58

 simultane

-0.57

 Koen

-0.57

POSITIVE LOGITS

anical

1.16

cham

0.99

anooga

0.97

ãĤ´ãĥ³

0.89

zeb

0.89

leneck

0.88

ania

0.88

herer

0.88

assium

0.87

wana

0.86

Activations Density 0.008%

mentions of the term "bot" in the text

mentions of robots or automated entities

No Comments

No Known Activations